TCA cycle involved enzymes SucA and Kgd, as well as MenD: Efficient biocatalysts for asymmetric C-C bond formation

Article abstract of DOI:10.1021/ol3031186

Asymmetric mixed carboligation reactions of α-ketoglutarate with different aldehydes were explored with the thiamine diphosphate dependent enzymes SucA from E. coli, Kgd from Mycobacterium tuberculosis, and MenD from E. coli. All three enzymes proved to be efficient biocatalysts to selectively deliver chiral δ-hydroxy-γ-keto acids with moderate to excellent stereoselectivity. The high regioselectivity is due to the preserved role of α-ketoglutarate as acyl donor for these enzyme-catalyzed reactions.

Full text of DOI:10.1021/ol3031186

ORGANIC
LETTERS
2013
Vol. 15, No. 3
452–455
TCA Cycle Involved Enzymes SucA and
Kgd, as well as MenD: Efficient Biocatalysts
for Asymmetric CÀC Bond Formation
Maryam Beigi,† Simon Waltzer,† Alexander Fries,† Lothar Eggeling,‡
§
,†
€
Georg A. Sprenger, and Michael Muller*
Institute of Pharmaceutical Sciences, Albert-Ludwigs-University of Freiburg,
Albertstraße 25, 79104 Freiburg, Germany, Institute of Bio- and Geoscience, IBG-1:
Biotechnology, Research Centre Ju€lich, 52425 Ju€lich, Germany, and Institute of
€
Microbiology, Universitat Stuttgart, Allmandring 31, 70569 Stuttgart, Germany
michael.mueller@pharmazie.uni-freiburg.de
Received November 12, 2012
ABSTRACT
Asymmetric mixed carboligation reactions of r-ketoglutarate with different aldehydes were explored with the thiamine diphosphate dependent
enzymes SucA from E. coli, Kgd from Mycobacterium tuberculosis, and MenD from E. coli. All three enzymes proved to be efficient biocatalysts to
selectively deliver chiral δ-hydroxy-γ-keto acids with moderate to excellent stereoselectivity. The high regioselectivity is due to the preserved role
of R-ketoglutarate as acyl donor for these enzyme-catalyzed reactions.
The tricarboxylic acid (TCA) cycle is a key component
of the central metabolism in aerobic organisms.1 Within
the TCA cycle, carbohydrates, fats, and proteins are catab-
olized into carbon dioxide and generate NADH, which
subsequently promotes oxidative phosphorylation to pro-
vide ATP as the source of usable energy. In the anabolic
function of the cycle, R-ketoglutarate (R-KG) is produced
as a precursor of glutamate, which is subsequently con-
verted into succinate (often through succinyl CoA). The
oxidative decarboxylation of R-KG to succinyl CoA is
known to be catalyzed by the R-ketoglutarate dehydrogen-
ase (KDH) enzyme complex.2 In principle, KDH is similar
to pyruvate dehydrogenase as it is composed of the follow-
ing three enzymes: thiamine diphosphate (ThDP) depen-
dent 2-oxo acid decarboxylase (E1), lipoate-dependent acyl
transferase (E2), and FAD-dependent dihydrolipoyl de-
hydrogenase (E3).3
The function of KDH starts with the decarboxylation of
R-KG by the E1 subunit, also designated as R-ketogluta-
rate decarboxylase, to form the ThDP adduct of succinic
semialdehyde (SSA). ThDP-dependent enzymes are known
to be involved in diverse ligase and lyase reactions.4 The
ThDP-dependent enzyme MenD (2-succinyl-5-enolpyruvyl-
6-hydroxy-3-cyclohexene-1-carboxylate synthase) from
E. coli K12 is a versatile biocatalyst for asymmetric syn-
thesis.5 MenD is involved in the second step of menaqui-
none biosynthesis and catalyzes a Stetter-like 1,4-addition
of R-KG toisochorismate.5d Like MenD, the E1 subunit of
(3) (a) Kanzaki, T.; Hayakawa, T.; Hamada, M.; Fukuyoshi, Y.;
Koike, M. J. Biol. Chem. 1969, 244, 1183. (b) Kubasik, N. P.; Richert,
D. A.; Bloom, R. J.; Hsu, R. Y.; Westerfeld, W. W. Biochemistry 1972,
11, 2225.
† Albert-Ludwigs-University of Freiburg.
€
(4) (a) Jordan, F. Nat. Prod. Rep. 2003, 20, 184. (b) Muller, M.;
‡ Research Centre J€ulich.
Gocke, D.; Pohl, M. FEBS J. 2009, 276, 2894. (c) Kluger, R.; Tittmann,
K. Chem. Rev. 2008, 108, 1797.
§
€
Universitat Stuttgart.
(1) Sweetlove, L. J.; Beard, K. F. M.; Nunes-Nesi, A.; Fernie, A. R.;
(5) (a) Popp, J. J. Bacteriol. 1989, 171, 4349. (b) Palaniappan, C.;
Sharma, V.; Hudspeth, M. E. S.; Meganathan, R. J. Bacteriol. 1992, 174,
8111. (c) Jiang, M.; Cao, Y.; Guo, Z. F.; Chen, M.; Chen, X.; Guo, Z.
Biochemistry 2007, 46, 10979. (d) Kurutsch, A.; Richter, M.; Brecht, V.;
Ratcliffe, R. G. Trends Plant Sci. 2010, 15, 462.
(2) (a) Tanaka, N.; Koike, K.; Hamada, M.; Otsuka, K. I.; Suematsu,
T.; Koike, M. J. Biochem. 1971, 69, 1143. (b) Bunik, V. I.; Strumilo, S.
Curr. Chem. Biol. 2009, 3, 279.
€
Sprenger, G. A.; Muller, M. J. Mol. Catal. B: Enzym. 2009, 61, 56.
r
10.1021/ol3031186
Published on Web 01/14/2013
2013 American Chemical Society

KDH also accepts R-KG as the physiological (donor)
substrate.
Furthermore, the 2,4-dinitrophenylhydrazones of the
reaction compounds were prepared at the starting point,
and at the end of the reaction, and were analyzed by HPLC.
NMR and HPLC assays both showed that the incubation of
R-KG in the presence of SucA and Kgd led to the release of
SSA in solution,12 while we were unable to detect any SSA
in the case of MenD. Nevertheless, the release of CO2 was
confirmed in the case of all three enzymes with NMR
experiments (using 13C-labeled R-KG).15
Next, the catalytic activity of all three enzymes with
various aldehydes using the physiological acyl donor
R-KG was investigated. The reaction conditions were opti-
mized for each enzyme using 2-fluorobenzaldehyde and
R-KG as a model reaction (see the Supporting Information).
All reactions were performed on an analytical scale (1.5 mL
reaction volume), and at least one example of each reaction
type was undertaken on a semipreparative scale (15 mL
reaction volume, 0.2 mmol acceptor substrate).
As indicated by the aromatic substrates, all three en-
zymes accepted a broad range of substituted benzalde-
hydes (Table 1, products 1À8). The electronic features of
the acceptor substrate have a direct impact on the reaction,
as the presence of an electron-withdrawing group such as
halide led to improved results while the presence of an
electron-donating group such as methoxy led to modest
results, except for the cases with MenD. The presence of a
nitro substituent at the ortho position resulted in no
product formation for all three enzymes and therefore is
not shown in Table 1. Substitution at different positions of
the aromatic ring does not seem to play a significant role
based on the results with chlorobenzaldehydes. SucA and
Kgd showed lower conversion and significantly decreased
enantioselectivity compared to MenD. As a result, MenD
is the preferred enzyme when aromatic acceptor aldehydes
are applied, resulting in R-configured products with good
to excellent regio- and enantioselectivity (76 to >99% ee
(enantiomeric excess)).
In a continuation of our studies on MenD,5d we were
interested in investigating whether SucA and Kgd,6 which
are known for their decarboxylation activity, can also
catalyze asymmetric CÀC bond formations. In the pres-
ence of an appropriate acceptor, chiral δ-hydroxy-γ-keto
acids can be synthesized as products. These are potent
precursors of γ- and δ-lactones which are present in the
structure of several natural products and are important
intermediates in organic synthesis.7,8 In this study, we
focused on mixed carboligation reactions of two different
carbonyl compounds using R-KG as one substrate.
The sucA gene in E. coli K129 encodes the E1 subunit of
KDH. It is a homologue of the odhA gene in Corynebac-
terium glutamicum10 and Bacillus subtilis.11 In Mycobac-
terium tuberculosis, the homologous protein is Kgd which
carries additional E2 sequences, as is also the case in
C. glutamicum. TheKgd proteinisessentialfor cellsurvival
of M. tuberculosis, but an additional KG:ferredoxin oxi-
doreductase may bypass the requirement for KDH activity
under anaerobic conditions.12,13
The SucA and Kgd recombinant proteins were designed
with a C-terminal hexa-histidine tag and were overex-
pressed in E. coli BL21(DE3) cells. Enzymes were purified
by immobilized metal chelate chromatography (Ni-NTA).
Production of MenD was performed as described pre-
viously.5d
In order to verify the decarboxylase activity of these
enzymes, the release of SSA from a solution of R-KG and
enzyme was followed. 13C-Labeled R-KG was synthesized
by oxidation of [1,2-13C]-L-glutamate using L-glutamate
dehydrogenase (L-glutamate-DH) from Clostridium sp.
coupled with NADH oxidase from Lactobacillus brevis.14
In situ biotransformation of the produced [1,2-13C]-R-KG
with each of the three enzymes (MenD, SucA and Kgd)
was then followed by 13C NMR spectroscopy (Scheme 1).
Further substrate studies were performed with a series of
aliphatic aldehydes with different chain length (Table 1,
products 9À12). In contrast to the aromatic aldehydes,
here, SucA and Kgd delivered better results than MenD,
especially in terms of enantioselectivity. In the case of
SucA, enantioselectivity dropped significantly when steri-
cally demanding aldehydes were used, whereas reverse re-
sults were obtained with MenD as the catalyst (<63% ee).
In the case of Kgd, there was no indication of an impact
of substrate steric demand on the ee. Overall, all three
Scheme 1. Synthesis and Use of 13C-Labeled R-KG by
Enzymatic Oxidation of L-Glutamate Coupled with
in Situ Decarboxylationa
a The 13C-labeled C atoms are marked with an asterisk.
(10) (a) Usuda, Y.; Tujimoto, N.; Abe, C.; Asakura, Y.; Kimura, E.;
Kawahara, Y.; Kurahashi, O.; Matsuil, H. Microbiology 1996, 142,
3347. (b) Hoffelder, M.; Raasch, K.; van Ooyen, J.; Eggeling, L.
J. Bacteriol. 2010, 19, 5203.
(11) Carlsson, P.; Hederstedt, L. J. Bacteriol. 1989, 171, 3667.
(12) Wagner, T.; Bellinzoni, M.; Wehenkel, A.; O’Hare, H. M.;
Alzari, P. M. Chem. Biol. 2011, 18, 1011.
(6) Schlossberg, M. A.; Bloom, R. J.; Richert, D. A.; Westerfeld,
W. W. Biochemistry 1970, 9, 1148.
(7) (a) Cavinato, G.; Toniolo, L.; Vavasori, A. J. Mol. Catal. 1994,
89, 93. (b) Cavinato, G.; Toniolo, L. J. Mol. Catal. 1993, 78, 121.
(8) (a) Yadav, J. S.; Mandal, S. S. Tetrahedron Lett. 2011, 52, 5747.
(b) Leßmann, T.; Leuenberger, M. G.; Menninger, S.; Lopez-Canet, M.;
ꢀ
(13) Baughn, A. D.; Garforth, S. J.; Vilcheze1, C.; Jacobs, W. R.
PLoS Pathog. 2009, 5, e1000662.
€
(14) (a) Odman, P.; Wellborn, W. B.; Bommarius, A. S. Tetrahedron:
€
€
Muller, O.; Hummer, S.; Bormann, J.; Korn, K.; Fava, E.; Zerial, M.;
Mayer, T. U.; Waldmann, H. Chem. Biol. 2007, 14, 443.
Asymmetry 2004, 15, 2933. (b) Geueke, B.; Riebel, B.; Hummel, W.
Enzyme Microb. Technol. 2003, 32, 205.
(15) These observations with purified MenD are in contrast to those
(9) (a) Darlison, M. G.; Spencer, M. E.; Guest, J. R. Eur. J. Biochem.
1984, 141, 351. (b) Frank, R. A. W.; Price, A. J.; Northrop, F. D.;
Perham, R. N.; Luisi, B. F. J. Mol. Biol. 2007, 368, 639.
previously reported using supernatant enzyme (ref 5b).
Org. Lett., Vol. 15, No. 3, 2013
453

Table 1. Conversion and ee Values for the Enzyme-Catalyzed
1,2-Addition of R-KG to Various Aldehydes
Scheme 2. Lactonization of γ-Keto Acid 13 to the
Corresponding δ-Lactone as a Possible Precursor for
the Anticancer Compound 5-Hydroxygoniothalamin
conversiona (%) (eeb (%)) of enzyme
R
product
SucA
Kgd
MenD
Ph
1
5 (6)
50 (31)
61 (41)
43 (57)
5 (nd)
9 (17)
45 (26)
50 (45)
57 (65)
50 (68)
2 (nd)
20 (52)
25 (20)
>99 (76)
>99 (82)
>99 (70)
>99 (72)
ncd
>99 (99)
>99 (94)
98 (93)
2-FC6H4
2-ClC6H4
2-BrC6H4
2-IC6H4
2
3
Table 2. Donor Substrate Spectra for the Enzyme-Catalyzed
1,2-Addition of Various R-Keto Acids to Aromatic Aldehydes
4
82 (80)
5
90 (76)
2-MeOC6H4
3-ClC6H4
6
35 (10)
50 (43)
45 (3)
98 (98)
7
>99 (96)
>99 (93)
>99 (<5)
>99 (11)
>99 (63)
>99 (61)
89 (nd)
4-ClC6H4
8
CH3
9c
10c
11c
12c
13
14
15
>99 (94)
>99 (94)
>99 (90)
>99 (82)
8 (nd)
CH3CH2
CH3(CH2)3
CH3(CH2)4
PhCHdCH
PhCHdC(CH3)
CH3CH2CHdCH
conversiona (%) (eeb (%)) of enzyme
R1
R2
product
SucA
Kgd
ncc
MenD
4 (nd)
nc
>99 (>99)
>99 (nd)
98 (nd)
H
F
H
F
F
F
CH3
CH3
17
18
19
20
21
22
26 (78)
37 (16)
32 (94)
52 (60)
nc
nc
nc
nc
nc
5 (nd)
nc
CH3CH2CHdC(CH3) 16
12 (nd)
nc
CH2CH3
nc
a Determined by NMR spectroscopy. b The absolute configuration
for all products was determined as (R) on the basis of circular dichroism.
c The conversion was determined to be quantitative as no aldehyde was
detected in the crude mixtures by NMR or GCÀMS analysis. d nc: no
conversion; nd: not determined.
CH2CH3
6 (nd)
nc
nc
CH2COOH
(CH2)3COOH
20 (nd)
nc
nc
nc
a Determined by GCÀMS. b Determined by chiral-phase HPLC (see
the Supporting Information). c nc: no conversion; nd: not determined.
enzymes gave excellent conversion for the formation of
δ-hydroxy-γ-keto acids from aliphatic aldehydes. SucA
and Kgd gave the expected products with moderate to
good enantioselectivity (70À94% ee).
Table 3. Conversion and ee Values for the Enzyme-Catalyzed
1,2-Addition of SSA to an Aromatic or Aliphatic Aldehyde
Next, the potential of different R,β-unsaturated alde-
hydes to act as acceptor substrates was evaluated. For this
type of substrate, addition of the decarboxylated R-KG
as the acyl donor can occur either at the carbonyl moiety in
a 1,2-addition or at the conjugated double bond, which
forms the Stetter product.16 As shown in Table 1 (products
13À16), Kgd and SucA led to no conversion or very poor
results while MenD gave the desired products (1,2-addition)
in excellent regio- and enantioselectivity. The formation of
1,4-addition products was not observed.
enzyme
Kgd
R
product
SucA
MenD
conversiona (%), (eeb (%))
CH3
Ph
17
1
quant (76)
1 (nd)
ncc
nc
20 (<5)
11 (99)
It is noteworthy to mention that products 13 and 14
can be considered as potential precursors of the styrenyl
δ-lactones which are observed in several bioactive com-
pounds, for example 5-hydroxygoniothalamin (Scheme 2).7b
In order to determine the acyl donor spectrum of all
three enzymes, different R-keto acids were tested as sub-
strates with benzaldehyde or 2-fluorobenzaldehyde as
acceptor. As shown in Table 2, MenD and Kgd are very
specific with respect to the donor substrate as they poorly
a Determined by NMR spectroscopy for 17 and by GCÀMS for 1.
b The absolute configuration for all products was determined as (R)
on the basis of circular dichroism. c nc: no conversion; nd: not
determined.
accept any other R-keto acid. In contrast, SucA showed a
broader donor substrate spectrum.
Furthermore, SSA was tested as a possible acyl donor,
and the results are shown in Table 3. Interestingly, SucA
and MenD showed a tolerance to accept SSA, whereas the
use of Kgd led tono productformation, independent of the
acceptor type.
€
€
(16) (a) Dresen, C.; Richter, M.; Pohl, M.; Ludeke, S.; Muller, M.
Angew. Chem., Int. Ed. 2010, 49, 6600. (b) Cosp, A.; Dresen, C.; Pohl,
€
€
M.; Walter, L.; Rohr, C.; Muller, M. Adv. Synth. Catal. 2008, 350, 759.
454
Org. Lett., Vol. 15, No. 3, 2013

revealed a hydrophobic pocket of MenD which is inter-
rupted by a histidine in MsKgd, a probable hotspot. Then
MsKgd and SucA were aligned on the basis of the primary
sequence. The resulting variant SucA-H460I (Fries and
Table 4. Carboligation Efficiency of SucA and the H460I
Variant
conversiona (%) (eeb (%))
€
Muller, to be published) led to excellent conversions and
enantioselectivities using aromatic substrates (Table 4).
Chemoselectivity, which is often a challenge in mixed
carboligation reactions,18 was excellent, independent of
the substrate or enzyme used. This is due to the preserved
role of R-KG as donor with these enzymes. Furthermore,
self-condensation of R-KG was never observed.
product
SucA
SucA-H460I
1
2
7 (6)
94 (93)
99 (83)
52 (31)
a Determined by GCÀMS. b The absolute configuration for all
products was determined as (R) on the basis of circular dichroism.
In addition, we have demonstrated that succinic semi-
aldehyde, even as a direct substrate, can be successfully
applied as a donor using ThDP-dependent enzymes. So far,
acetaldehyde (pyruvate), glyoxal (hydroxypyruvate), and
aliphatic aldehydes have been shown to be the preferred
donor substrates for ThDP-dependent enzymes.4b There-
fore, our results expand the range of substituted acetalde-
hydes (glyoxal, and methoxy- and dimethoxyacetaldehyde)
as substrates toward C4-functionalized aldehydes.
In summary, we have characterized the substrate spectra
of enzymes from the central metabolism of aerobic
organisms, namely SucA from E. coli and Kgd from
M. tuberculosis, in comparison with that of MenD from
E. coli. It was shown that these three enzymes feature a
broad, but different, substrate specificity for cross-acyloin
condensations. It was demonstrated that a wide range of
substrates, including aliphatic, aromatic and R,β-unsaturated
aldehydes, in combination with R-KG can be used, resulting
in the expected δ-hydroxy-γ-keto acids.
The products are formed with moderate to excellent
enantioselectivity. In order to enhance the efficiency of
these selected enzymes, protein engineering can be applied
to prepare variants with improved carboligation efficiency
or showing S-selectivity.17 To identify the critical residues
responsible for the differences in the activity, we compared
the structure of MenD (PDB entry 3HWX) with the
structure of Kgd from Mycobacterium smegmatis
(MsKgd, PDB entry 2YID). The overlay of the active sites
Acknowledgment. We thank Volker Brecht and Sascha
Ferlaino, University of Freiburg, for NMR measurements,
€
Prof. Werner Hummel, Research Centre Julich, for providing
L-glutamate-DH and NADH oxidase, and Dr. Loay Al-
Momani, Tafila Technical University, for his useful com-
ments. We thank the German Research Foundation DFG for
financial support within the framework of project FOR 1296.
Supporting Information Available. Detailed experimen-
tal procedures and characterization data of all new
compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
€
(17) (a) Meyer, D.; Walter, L.; Kolter, G.; Pohl, M.; Muller, M.;
€
(18) (a) Piel, I.; Pawelczyk, M. D.; Hirano, K.; Frohlich, R.; Glorius,
Tittmann, K. J. Am. Chem. Soc. 2011, 133, 3609. (b) Gocke, D.; Kolter,
G.; Gerhards, T.; Bertold, C. L.; Gauchenova, E.; Knoll, M.; Pleiss, J.;
Muller, M.; Schneider, G.; Pohl, M. ChemCatChem 2011, 3, 1587. (c)
Gocke, D.; Walter, L.; Gauchenova, E.; Kolter, G.; Knoll, M.; Berthold,
F. Eur. J. Org. Chem. 2011, 5475. (b) Rose, C. A.; Gundala, S.; Connon,
S. J.; Zeitler, K. Synthesis 2011, 190. (c) O’Toole, S.; Rose, C. A.;
Gundala, S.; Zeitler, K.; Connon, S. J. J. Org. Chem. 2011, 76, 347.
€
€
C. L.; Schneider, G.; Pleiss, J.; Muller, M.; Pohl, M. ChemBioChem
2008, 9, 406. (d) Reetz, M. T. Angew. Chem., Int. Ed. 2011, 50, 138.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 3, 2013
455